Positive inotropic drugs: Cardiac glycosides (digoxin)

Digoxin is a member of a class of drugs known as the cardiac glycosides that also includes digitoxin and ouabain. Cardiac glycosides occur naturally in plants of the genera Digitalis, such as foxgloves and Strophanthus. Only digoxin and very rarely digitoxin are used clinically. Such agents increase the force of contraction of the heart, a positive inotropic action which underlies their use in some cases of heart failure. They also have important effects upon electrical conduction in the heart, particularly the velocity at which the action potential is conducted through the atrioventricular (AV) node. Clinically, digoxin is used in severe heart failure, but usually as a third line agent when other treatments including ACE inhibitors, low dose β-blockers and aldosterone receptor antagonists, or angiotensin receptor blockers do not provide sufficient benefit. However, as explained below, digoxin may be introduced at an earlier stage if the patient has atrial flutter as a co-morbidity.

Relevant Chemistry

Structurally, the cardiac glycosides considered here (the cardenolides) consist of a steroid ring to which a lactone and sugar residues are attached (in the β-configuration) at the C17 (D ring) and C3 (A ring) positions, respectively. An unsaturated lactone ring is essential for the pharmacodynamic action of the cardiac glycosides, as is cis-fusion of the steroid A/B and C/D rings, trans-fusion of the B/C rings and the presence of a β-hydroxyl at C14. The sugar groups (which are variable in nature and number, e.g. one rhamnose in ouabain, or three digitoxose in digoxin and digitoxin) influence the potency and pharmacokinetics of individual compounds. The nature of the sugar also contributes to the modest selectivity of cardiac glycosides between isoforms at their primary molecular target, the Na⁺/K⁺-ATPase. The latter is classically regarded as a heterodimer of α (α1, α2, α3 and α4) and β (β1, β2 and β3) subunits in a 1:1 stoichiometry in association with regulatory subunits (phospholemman in the heart)

Mechanism of action

Cardiac glycosides bind to the catalytic α-subunit of the Na⁺/K⁺-ATPase (the ‘sodium pump’) inhibiting its action to transport Na⁺ out of, and K⁺ into, the cardiac muscle cell. Thus, therapeutic concentrations of digoxin bind to a proportion of the Na⁺/K⁺-ATPase pumps in cardiac muscle, reducing overall pumping activity. Excessive inhibition of the pump underlies many of the serious adverse effects of digoxin which limit its use (see below).

Binding of digoxin occurs at the extracellular side of the pump in competition with K⁺ explaining, at least partially, the clinically important phenomenon that a reduced concentration of K⁺ in the plasma (hypokalaemia) increases the action of digoxin which may precipitate serious toxicity. In addition, reduced plasma K⁺ may result in the phosphorylation of the Na⁺/K⁺-ATPase increasing its affinity for binding of digoxin and thus pump occupancy. Both α1β and α2β isoforms of the Na⁺/K⁺-ATPase have been implicated in the inotropic action of cardiac glycosides.

In cardiac muscle, which due to its high level of electrical activity is particularly reliant upon the Na⁺/K⁺-ATPase to maintain appropriate ion gradients across the plasma membrane, inhibition of the pump causes an elevation of the intracellular concentration of Na⁺ ([Na⁺]_i). This is accompanied by a small reduction in the resting (diastolic) membrane potential of cardiac muscle cells because the pump is electrogenic (i.e. pumps 3 Na⁺ out: 2 K⁺ in for each transport cycle at the expense of one molecule of ATP hydrolysed to ADP and P_i). The subsequent reduction in the electrochemical gradient for Na⁺ entry secondarily reduces the expulsion of Ca²⁺ from the cytoplasm during diastole by the plasma membrane Na⁺/Ca²⁺ exchanger (NCX1) operating with the stoichiometry 3 Na⁺ in: 1 Ca²⁺ out during each transport cycle. This occurs because the operation of the Na⁺/K⁺-ATPase (primary active transport) is required to maintain the secondary active transport mediated by NCX1. The excess Ca²⁺ is sequestered into the lumen of the sarcoplasmic reticulum (SR) by the Ca²⁺-ATPase (SERCA2a) within the membrane of that organelle. Thus, additional free Ca²⁺ is available for release from the SR lumen during the plateau phase (phase 2) of the ventricular action potential by the process of calcium-induced calcium release (CICR). The latter process is elaborated upon in the textbox. The cytoplasmic Ca²⁺ transient that ensues and generates systole is thus elevated and via increased occupation of the cardiac isoform of troponin-C (TNNC1) by Ca²⁺ translates into an increase in cardiac contractility.

The effect of digoxin upon the electrical activity of the heart is complex and consists of a number of direct and indirect actions. Directly, inhibition of the Na⁺/K⁺-ATPase causes a small depolarization of cardiac muscle (see above) that predisposes to abnormal discharge of action potentials by cardiac muscle cells and consequently ectopic beats (see below). Acting directly upon the conduction system of the heart, digoxin slows conduction velocity through the atrioventricular (AV) node by prolonging the refractory period. The latter action is useful and underlies the use of the drug in patients with atrial fibrillation or flutter accompanied by a high ventricular rate, usually if combined with heart failure. Although digoxin does not correct atrial fibrillation or flutter, it suppresses aberrant impulses from conducting through the AV node to trigger ventricular arrhythmias. Conversely, excessive suppression of AV conduction leads to heart block, one of the characteristic features of digoxin toxicity. Indirectly, but again through inhibition of the Na⁺/K⁺-ATPase, digoxin stimulates the parasympathetic nerve supply to the heart by acting upon the central vagal nucleus (i.e. increases vagal tone) which reinforces its suppressant effect upon AV node conduction velocity and also decreases the automaticity of the sinoatrial (SA) node, thus reducing discharge rate. The latter action slows the heart rate in sinus rhythm.

Pharmacokinetics

Digoxin may be administered by mouth, or if a rapid action is required, by slow intravenous injection. When given by mouth, the bioavailability of digoxin is influenced by the formulation of the drug (tablets, or liquid). The drug has a high apparent volume of distribution, principally due to binding to the Na⁺/K⁺-ATPase of skeletal muscle, and coupled with an elimination half-life of approximately 36 hours (with normal renal function, see below) this necessitates a loading dose usually by mouth if a rapid action is required. The drug may be taken with, or without, food.

Digoxin is a polar molecule and the main route of elimination (approximately 70% of the drug unchanged) is via renal excretion involving both glomerular filtration and active tubular secretion. Thus, in significant renal impairment, the dose of digoxin must be reduced particularly in view of the very low therapeutic index of the drug. The reduction in dosage is proportional to the fall in glomerular filtration rate (GFR). Hepatic metabolism of digoxin contributes to elimination to a much lower extent than renal excretion.

Drug interactions

The margin of safety between the doses of digoxin that produce a therapeutic, or toxic, effect is very narrow (plasma concentrations of approximately 1 – 2.6 nmol/l are reported as the therapeutic window). Moreover, digoxin exhibits important pharmacodynamic and pharmacokinetic interactions with many drugs, including some used to treat heart disease. Some important examples are:

Pharmacodynamic – these are largely predictable from a basic understanding of cardiovascular pharmacology:

• with β-adrenoreceptor antagonists: AV node conduction velocity is facilitated by noradrenaline (and adrenaline) acting as agonists upon β₁-adrenoceptors in nodal tissue. Blockade of this effect by β₁-antagonists, in combination with the negative action of digoxin upon AV node conduction velocity, can lead to a high grade AV block. In addition, β-blockers exert a negative inotropic action that physiologically antagonises the inotropic action of digoxin.
• with Ca²⁺ channel blockers: Ca²⁺ channel blockers that exhibit a degree of selectivity for L-type Ca²⁺ channels in cardiac muscle (e.g. verapamil) can produce a negative inotropic action, negating the beneficial positive inotropic action of digoxin.
• with thiazide and loop diuretics: both classes cause loss of potassium from the body leading to hypokalaemia. As explained above, this potentiates the action of digoxin leading to potential toxicity.

Pharmacokinetic – these are less easily understood from a knowledge of mechanism and are probably best consulted in comprehensive textbooks of basic and clinical pharmacology and national and local formularies:

• with drugs that increase the absorption of digoxin from the gastrointestinal tract. Many antibiotics (e.g. erythromycin) may cause this by depleting intestinal flora that metabolise a fraction of the drug before it is absorbed across the mucosa.
• with drugs that alter the apparent volume of distribution, or renal clearance, of digoxin. Examples of particular relevance due to their cardiovascular/renal actions are verapamil, quinidine, amiodarone and spironolactone.

Adverse effects

The major adverse effects of digoxin include:

• bradycardia
• block of AV node conduction (see above): digoxin is contraindicated in patients with second degree heart block, or intermittent complete heart block
• triggering a variety of arrhythmias including ectopic beats (see above): delayed after depolarizations (DADs) may result from Ca²⁺ overload causing coupled beats (bigeminy)

Digoxin is contraindicated in patients who have, or are at risk of

• ventricular arrhythmias
• gastrointestinal disturbances such as anorexia, nausea, vomiting and diarrhoea
• neurological disturbances such as yellow vision (probably due to an action on the retina), fatigue, malaise and confusion.

Digoxin toxicity may be precipitated by electrolyte disturbances that include hypokalaemia (see above), hypomagnesaemia and hypercalcaemia. Treatment of digoxin toxicity includes withholding the drug, potassium supplementation to correct hypokalaemia and in severe cases the administration of digoxin specific antibody fragments.